The invention relates to magnetic recording and reproducing apparatuses such as home VTRs in which a luminance signal of a video signal to be recorded is subjected to frequency modulation and then to single-sideband recording.
In home VTRs the frequency band of signals which can be recorded and reproduced is generally narrower than the frequency band of a frequency-modulated (FM) luminance signal of a video signal to be recorded and, as a result, high-frequency components of the upper sideband of this FM signal are neither recorded nor reproduced. It is for this reason that the FM signal takes the form of a single-sideband wave when recorded and involves an amplitude variation when reproduced. If this FM signal is demodulated with its amplitude variation removed by a limiter, the aforesaid high frequencies of its upper sideband can be recovered This is how the single-sideband recording is performed.
By the way, in the FM signal that has been through with the limiter, the modulation indices of the high-frequency components of its recovered upper sideband and the low-frequency components of its lower sideband are decreased, and this acts to decrease the level of the FM signal. When this FM signal is demodulated, the energy of the high-frequency of the signal level, the home VTR employs a peaking circuit to boost the high-frequency components of the demodulated luminance signal. This peaking circuit serves to keep the level of the high-frequency components of the demodulated luminance signal as high as that of its low-frequency components.
Further, video tapes for the home VTRs include normal tapes and so-called "high-grade tapes" that exhibit higher performance than the normal tapes. To make the most of the performance of high-grade tapes, the characteristics of a recording or reproducing system of a VTR must be differentiated between high-grade tapes and normal tapes. For example, the following means could be proposed to implement such differentiation.
(1) Controlling the level of emphasis at a detail emphasis circuit in a recording system in which a luminance signal of a video signal is frequency-modulated and then magnetically recorded. The detail emphasis circuit serves to emphasize small-level signal components in the high-frequency components of the luminance signal before frequency modulation; PA0 (2) Controlling the amount of equalization at an FM equalizer which serves to boost the low-frequency components of an FM luminance signal in the recording system in which the luminance signal of a video signal is frequency-modulated and then magnetically recorded; PA0 (3) Controlling the amount of peaking at an FM peaking circuit which serves to boost the high-frequency components of an FM luminance signal before demodulation in a reproducing system in which a reproduced luminance signal is frequency-demodulated; and PA0 (4) Controlling the amount of cancellation at a noise canceller which serves to cancel noise of a demodulated luminance signal in the reproducing system in which a reproduced luminance signal is frequency-demodulated. PA0 (1) To identify the type of a video tape by the level of a reproduced signal, the frequency characteristic of a recording system is adjusted in such a manner that a reproduced level is detected and stored and the frequency characteristic of an FM luminance signal is adjusted in accordance with the stored reproduced level (Japanese Patent Unexamined Publication No. 146674/1988); PA0 (2) The amount of compensation for a frequency characteristic is kept unchanged by changing the peaking amount at an FM equalizer in the reproducing system while detecting the envelope of a reproduced FM signal and using this detected output as a control signal, even if the frequency o characteristic of a video tape has been changed (Japanese Patent Unexamined Publication No. 59287/1988); and PA0 (3) A noise component from an original signal is subtracted by separating high-frequency components in the reproduced luminance signal at which noise tends to concentrate by a filter, i.e., the noise canceller 25 described as being included in the reproducing system, and by further filtering the low-level signal components out by the slice circuit, so that the high-frequency low-level signal is considered as the noise and subtracted from the original signal; that is,
The recording and reproducing systems of a home VTR will be outlined next with reference to FIG. 3
The recording system will be described first. In FIG. 3, an input terminal 1 receives a video signal from a TV receiver or the like (not shown). This video signal is applied to a LPF 2, where a luminance signal is extracted, and the extracted luminance signal is then applied to an AGC 3, which controls the applied signal in such a manner that the level of its synchronizing signal can be kept constant. The video signal fed to the input terminal 1 is also applied to a BPF (not shown), where a chroma signal is extracted, and the extracted chroma signal is then converted into a low-frequency signal at a chroma signal processing circuit (not shown).
The luminance signal which has been through with the AGC 3 is then fed to a detail emphasis circuit 4, where small-level signal components in high-frequency components of the luminance signal are emphasized, and the emphasized signal is then applied to a luminance signal processing circuit 5. In the luminance signal processing circuit 5, the high frequencies are subjected to emphasis and a like process and then applied to a frequency modulator 6. The luminance signal frequency-modulated at the frequency modulator 6 has its low frequencies boosted by an FM equalizer 7, and low frequencies that correspond to the band of the low-frequency converted chroma signal are cut off at a HPF 8. The FM luminance signal which has been through with the HPF 8 is then applied to a recording amplifier 9, where it is amplified while mixed with the low-frequency converted chroma signal from the input terminal 10, and the thus processed signal is then supplied to a video head 11 to be magnetically recorded on a video tape 12.
The reproducing system will be described next. A video head 13, though shown separately from the video head 11 for purposes of convenience, is the same as the video head 11. A video signal reproduced from the video head 13 is amplified by a preamplifier 14, and not only a chroma signal that is subjected to low-frequency conversion by a LPF 15 is extracted but also a luminance signal that is frequency-modulated by a HPF 16 is extracted. The chroma signal which has been through with the LPF 15 is converted into a chroma signal of a subcarrier band by a chroma signal processing circuit 17 and the converted chroma signal is then supplied to a mixing circuit 18.
The luminance signal which has been through with the HPF 16 has its high-frequency components (around 5 MHz) boosted by an FM equalizer 19, is kept at a constant level by an FM AGC 20 thereafter, and is then demodulated when applied to a frequency demodulator 21. The luminance signal delivered from the FM demodulator 21 is appropriately processed by a luminance signal processing circuit 22, and has its high-frequency components (2 MHz or more) boosted by a subsequent peaking circuit 23. The luminance signal which has been through with the peaking circuit 23 is applied to a line noise canceller 24.
As shown in FIG. 4 in detail, the line noise canceller 24 delays a luminance signal fed from an input terminal 27 via the peaking circuit 23 for a single horizontal scanning period (hereinafter referred to simply as "1H") by a 1H delay element 28. The thus delayed luminance signal and the original luminance signal that has not been delayed are subjected to subtraction at a subtractor (which is shown as an adder 29, but acts as a subtractor because its polarity is inverted by the 1H delay element 28). Only a small-level signal which can be deemed as a noise component is extracted from the subtracted signal by a limiter 30, and the level of this small-level signal is attenuated by 1/2 at an attenuator 31. The attenuated signal is then subtracted by a subtractor shown as an adder 32, so that the noise component can be canceled. The luminance signal whose noise component has been canceled is then fed to a noise canceller 25 from an output terminal 33.
The noise canceller 25 extracts high-frequency components containing much noise from the luminance signal that has been through with the line noise canceller 24, and components whose levels are greater than a predetermined level are removed from the high-frequency components by a slice circuit (not shown). The output of the slice circuit is then subtracted from the luminance signal that has been through with the line noise canceller 24 to cancel the noise.
The luminance signal which has been through with the noise canceller 25 is applied to the mixing circuit 18 so as to be mixed with the chroma signal, the mixed signal is then delivered to an output terminal 26 and fed to a not shown well known CRT or the like, so that a reproduced image can be obtained.
The so-called high-grade tapes are commercially available as video tapes for home VTRs. The high-grade tape can provide not only a high reproducing output from the tape but also a remarkably improved signal-to-noise ratio by its extremely high degree of granulation and loading of magnetic powder.
That is, as shown in a characteristic diagram in FIG. 5, a high-grade tape a1 exhibits an output level higher than a normal tape b1, with the characteristic that its output level becomes higher especially in a high-frequency range (4 to 6 MHz) than in a low-frequency range (1 to 2 MHz).
When a video signal is recorded/reproduced using the high-grade tape, the levels of components such as the carrier wave, the low-frequency components of the upper sideband, and the high-frequency components of the lower sideband in the reproduced FM signal are boosted, making the levels of components such as the carrier wave, the low-frequency components of the upper sideband, and the high-frequency components of the lower sideband in the reproduced FM signal that has been through with the limiter 30 higher than in the normal tape. Also, the levels of components such as the high-frequency components of the upper sideband and the low-frequency components of the lower sideband in the reproduced FM signal that has been through with the limiter 30 become relatively lower than the levels of the low-frequency components of the upper sideband and the high-frequency components of the lower sideband as described before. And this makes the high-frequency signal level of the demodulated luminance signal low with respect to its low-frequency signal level that is maintained at a predetermined level, with this high-frequency signal level being lower than that in the normal tape. When the high-frequency signal level is so low as this, it cannot be adequately compensated for by the peaking circuit 23 if the circuit 23 is set to the normal tape mode.
FIG. 6 shows levels of demodulated video signals for both a high-grade tape a2 and a normal tape b2, indicating that the high-grade tape outputs a lower signal level at high frequencies than the normal tape. It is well known that high-grade tapes produce low reproduced outputs at high frequencies, and this results in blurring reproduced images.
To compensate for the impaired reproduction due to blurred images which are caused by deterioration in the high frequencies of a luminance signal of a recorded/reproduced video signal when operating a home VTR, the detail emphasis circuit 4, which serves to emphasize low-level signal components in high frequencies of a luminance signal to be recorded is provided as described above
As shown in FIG. 13, the aforesaid detail emphasis circuit 4 is made up of a HPF 78, an amplifier 79, a limiter 80, an attenuator 81, an adder 82, and the like. A luminance signal received at an input terminal 70 from the AGC 3 is subjected to a process so that its high-frequency components are extracted by the HPF 78, and the extracted components are then amplified by the amplifier 79. The amplified high-frequency components are then applied to the limiter 80 so that the small-level signal components are extracted. The high-frequency small-level signal components which have been through with the limiter 80 are subjected to level adjustment by the attenuator 81 and to addition thereafter by the adder 82 so as to be added to the luminance signal from the input terminal 70. Then, the thus processed signal is applied to the luminance signal processing circuit 5 via an output terminal 84.
In a VTR having this detail emphasis circuit 4, if it assumed that a high-frequency small-level signal component incorporated in the luminance signal of a video signal to be recorded, i.e., an input detail signal, is such as one shown by b3 in FIG. 15, then this input detail signal b3 is subjected to emphasis by the detail emphasis circuit 4 in the normal tape mode so that it will become a signal such as shown by b4 in FIG. 15. When the thus processed input detail signal b4 is recorded/reproduced by a normal tape, and further demodulated, a signal similar to the input detail signal b3 is reproduced as shown by b5 in FIG. 15.
By the way, if an input detail signal a3 shown in FIG. 15 which is similar to the input detail signal b3 is subjected to emphasis by the detail emphasis circuit 4 in the normal tape mode so that it will be reformed into a signal a4 that is similar to the signal b4, and if this signal a4 is recorded/reproduced using a high-grade tape and further demodulated, then the level of this signal a4 becomes lower than that of the input detail signal a3 as shown by a signal a5 in FIG. 15. Such decrease in the level of the detail signal results in blurs on reproduced images.
As described above, the amount of emphasis in the conventional detail emphasis circuit is not sufficient to compensate for the blurs on the reproduced images when high-grade tapes are used.
Further, to automatically provide the optimal performance of a video tape in accordance with its type as well as to automatically compensate for deterioration in the characteristic of a video tape in accordance with its type, the following means have heretofore been known.
an FM luminance signal reproduced from a video tape is detected using an FM reproducing level detector and the slice level of the slice circuit is decreased with increasing reproduced levels (e.g., in the case of a high-grade tape) (Japanese Patent Unexamined Publication No. 14178/1989).
Incidentally, since the noise canceller 25 at which the slice level remains constant regularly cancels any high-frequency low-level signal components, the original signal components other than the noise may also be canceled. For this reason, it is preferable to set the amount of cancellation to a small value for high-grade tapes in which the signal-to-noise ratio is significantly improved but the high-frequency level of their demodulated luminance signal is decreased.
The line noise canceller shown in FIG. 4 does cancel not only noise components but also small-level signal components (detail signals) to a certain degree, thus causing the level of small-level signals to be decreased.
Further, while there is no attenuation in the level of a luminance signal that is in identical correlation with a luminance signal of 1H before, once there is any deviation from the correlation, the level of the deviating signal is decreased as much as such deviation.
The cases where the correlation is identical and where there is a deviation in the correlation will be described with reference to FIGS. 7 and 8. Signals Xl to X5 in FIG. 7 show a case of the identical correlation, where Xl is an original signal; and X2, a signal of 1H before. These signals are in phase. Signal X3 is an output of the adder 29, which is set to level 0; X4, an output of the attenuator 31, which is likewise set to level 0; and X5, an output of the adder 32, producing a signal identical to the original signal shown by X1. Signals Y1 to Y5 show a case where a signal of 1H before is 90.degree. ahead of an original signal, with the original signal and the signal of 1H before being assumed to be such small-level signals as not to be limited by the limiter 30.
Signal Y1 is an original signal, expressed by sin.theta.; Y2, a signal of 1H before, expressed by sin(.theta.+ .pi./2), or sin(.theta. + .pi./2) = cos.theta.. Signal Y3 is an output of the adder 29, which is expressed by Y1-Y2=sin.theta.-cos.theta., or sin.theta.-cos.theta.=.sqroot.2 sin(.theta.-.pi./4) in a composite equation. Signal Y4 is an output of the attenuator 31, which is 1/2 the level shown by signal Y3, or .sqroot.2/2 sin(.theta.-.pi./4). Signal Y5 is an output of the adder 32, which is sin.theta.-1/2 [sin.theta.-cos.theta. ] =1/2 sin.theta.+1/2 cos.theta., or 1/.sqroot.2 sin(.theta.+.pi./4) in a composite equation.
As described above, the level of the original signal Y1 is decreased by 30% at the stage of the output signal Y5 and the phase of signal Y5 is 45.degree. ahead.
Small-level signals are generally less correlative in terms of line compared with large-amplitude signals. While no images can be recognized with the large amplitude signals having no line-based correlation, the small-level signals can produce recognizable images with no such correlation. For example, in the case of irregularities of human skin, there is no correlation. Signals Z1 to Z5 shown in FIG. 8 present a case where there is no correlation between an original signal and a signal of 1H before. In FIG. 8, signal Z1 is a projecting original signal; Z2, a flat signal of 1H before, these signals having such small-amplitudes as not to be limited by the limiter 30; Z3, an output of the adder 29, which has a projecting waveform identical with the original signal; Z4, an output of the attenuator 31, whose amplitude is 1/2 that of the original signal Z1; and Z5, an output of the adder 32, whose amplitude is 1/2 that of the original signal Z1.
Accordingly, while serving to reduce noise, the line noise canceller 24 also decreases the level of the detail signal to some extent, and if there is a deviation from the correlation between the detail signal and the luminance signal of 1H before, the line noise canceller decreases the level of the detail signal as much as that deviation
By the way, the aforesaid means have heretofore been known to automatically provide an optimal performance of a video tape in accordance with its type such as a normal tape or a high-grade tape, as well as to automatically compensate for deterioration in the characteristic of a video tape in accordance with its type. However, these means are not successful in effectively providing the optimal performance and compensating for the deterioration in the characteristic in accordance with the type of a video tape.
Let us now take a look at a case where a noise canceller disclosed in Japanese Patent Unexamined Publication No. 14178/1989 is used.
It is generally known that, when a color video signal demodulated thereafter, an unwanted component of about 1.2 MHz is present in the demodulated luminance signal. The level of this unwanted component is increased with higher level of a reproduced chroma signal, i.e., higher degree of color saturation.
Some reasons why this unwanted component is present will be described. In the case of magnetic recording on a VTR, a recording current is subjected to third order distortion by hysteresis and magnetized as such on the tape.
It is assumed that:
FM luminance signal = A sin.alpha.
Low-frequency converted chroma signal = B sin.beta.,
then, the above phenomenon on the tape can be expressed as follows. ##EQU1##
FIG. 9 is a spectrum obtained when a signal magnetized on a video tape has been reproduced. Reference character a designates a low-frequency converted chroma signal sin.sup.3 .beta.; b, an unwanted component sin(.alpha.-2.beta.); c, an FM luminance signal sin.sup.3 .alpha.; d, an unwanted component sin(.alpha. + 2.beta.); and e, f, unwanted components sin(2.alpha.-.beta.), sin(2.alpha.+.beta.), respectively. Among these unwanted components, the high-frequency components e, f produce so small output levels that they can be considered negligible.
In the case where noise of a luminance signal reproduced from a normal tape is to be canceled by the noise canceller whose amount of cancellation is varied in accordance with the level of the reproduced FM luminance signal as described above, the slice level is increased, so that the unwanted components b, d will also be canceled
However, in the case where noise of a luminance signal reproduced from a high-grade tape is to be canceled and where an image such as an animation whose color saturation is high is to be reproduced, the unwanted components b, d will not be canceled despite the fact that the slice level is small, because these components b, d are comparatively large.
The presence of the unwanted components b, d in a luminance signal causes mesh-like noise to appear over the entire part of a reproduced image, making the image indistinct.
In the case of an image with a low color saturation, the levels of the unwanted components b, d are low even if a luminance signal is derived from a high-grade tape, thereby causing no mesh noise to appear.